About the Role
This role involves creating and fine-tuning AI models to generate consistent, stylized characters using LoRA technology, with full remote work based in Spain.
Responsibilities
- Design original character concepts aligned with project requirements
- Generate high-quality training datasets for LoRA fine-tuning
- Train and refine LoRA models using diffusion-based AI systems
- Optimize character consistency across multiple poses and expressions
- Collaborate with engineers to integrate models into production pipelines
- Maintain documentation for training processes and model versions
- Iterate on feedback from creative and technical stakeholders
- Ensure outputs meet ethical and safety guidelines
- Stay current with advancements in generative AI and character modeling
- Troubleshoot model performance issues and adjust training parameters
- Produce diverse character styles while preserving artistic coherence
- Balance creative direction with technical constraints
- Contribute to best practices for AI-generated character development
- Participate in team reviews and knowledge-sharing sessions
- Support quality assurance for generated outputs
Nice to Have
- Experience with 3D-to-2D character projection workflows
- Familiarity with animation or rigging pipelines
- Knowledge of facial expression modeling in AI systems
- Previous work in gaming or entertainment industries
- Experience mentoring others in AI art techniques
- Contributions to open-source AI art projects
- Understanding of copyright and IP considerations in training data
Compensation
Competitive salary based on experience
Work Arrangement
Full remote, based in Spain
Team
Collaborative AI research and development team focused on generative models
Creative Freedom
You will have the opportunity to shape the visual identity of AI-generated characters while working within defined artistic guidelines.
Tech Stack
- Primary tools include Stable Diffusion, custom training scripts, and internal evaluation platforms.
- Familiarity with Python and command-line training interfaces is beneficial.
Not applicable
